Region of interest
It is possible to work on only one section of the image with an ROI. For this, the user can specify the area he wants in the configuration file. At the moment, our only option is to work with a sensor ROI.
Configuration and parameters
Name |
Description |
Type |
Default value |
Required |
|---|---|---|---|---|
col |
Columns sensor coordinates |
dict |
None |
No |
row |
Rows sensor coordinates |
dict |
None |
No |
Row and col parameters contain dictionaries, each is different depending on the mode used.
Sensor mode : The user uses pixel coordinates to select an ROI.
Mode |
Name |
Description |
Type |
Default value |
Required |
|---|---|---|---|---|---|
sensor |
first |
First pixel in selected coordinates |
int |
Yes, if ROI is desired. |
|
sensor |
last |
Last pixel in selected coordinates |
int |
Yes, if ROI is desired. |
Note
It is necessary for the first coordinate to be lower than the last coordinate.
Warning
If the input disparities are specified as paths to a pandora2d output folder containing disparity maps smaller than the image, these maps will be used as the ROI. The origin point and the step used to calculate the disparity maps are retrieved in order to correctly reconstruct the corresponding ROI.
If, in this case, a user ROI has already been specified, it will be overwritten by the one constructed from the disparity maps.
Example
{
"input":
{
// input content
},
"ROI":
{
"col": {"first": 10, "last": 100},
"row": {"first": 10, "last": 100}
},
"pipeline" :
{
// pipeline content
},
"output" :
{
// output content
}
}